gimble-workflows

Pass

Audited by Gen Agent Trust Hub on Sep 23, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
  • [COMMAND_EXECUTION]: The skill documents framework primitives for shell execution, including RunCommand, Check, and Service. It specifies that the Service function executes strings via zsh -c to manage background dependencies.
  • [INDIRECT_PROMPT_INJECTION]: The workflow patterns describe agents ingesting research findings and worker feedback. The skill provides safety guidance for these ingestion points, including the use of independent validators, explicit handoff boundaries, and the requirement that agent prose cannot override command execution evidence.
  • [NO_CODE]: The skill consists entirely of instructional markdown files and reference patterns. It contains no executable scripts, binaries, or dynamic configuration files that perform logic during skill operation.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 23, 2026, 11:38 PM